Speech delay of unknown origin (SD) (Shriberg, 1980) is a risk factor for a child's literacy achievement, positive self-concept, peer acceptance, and vocational choices. The high prevalence of SD (3.8% at 6 years of age; Shriberg, Tomblin, & McSweeny, 1999) places it among the most frequently occurring childhood disorders requiring public health resources for research and long-term treatment. The present research program has developed an etiological classification system for SD that proposes to explicate the distal causes for over 90% of children referred for treatment (Shriberg & Kwiatkowski, 1994). Specific aims for the next period of study are to cross-validate the accuracy of phenotype and diagnostic markers for each of three subtypes of SD and to describe associated speech acquisition processes. Seven large, demographically diverse databases will be used to complete six study series including data from over 2,200 children with SD. Project I: Speech-Genetics Studies. Approximately 60% of children with SD are posited to have a genetic subtype inherited as a quantitative (i.e., non-Mendelian) trait. Study Series 1 and 2 will cross-validate two perceptual and two acoustic phenotype markers for probands and family members. Study Series 3 will cross-validate prior familial aggregation findings, and Study Series 4 will explicate short-term and long-term normalization processes in this subtype of SD. Project Ih Speech-Otitis Media Studies. Approximately 30% of children with SD are posited to have histories of fluctuant hearing loss associated with recurrent otitis media with effusion. Research indicates that this proposed etiological subtype of SD is clinically under-diagnosed (i.e., false negatives). Study Series 5 will cross-validate a perceptual and an acoustic diagnostic marker for this subtype of SD. Project IIh Speech-Apraxia Studies. Approximately 1-2 children per 1000 are posited to have a subtype of SD that reflects the speech motor constraints associated with a disorder of speech praxis. Research indicates that this proposed etiological subtype of SD is clinically over-diagnosed (i.e., false positives). Study Series 6 will cross-validate two acoustic diagnostic markers for this subtype of SD.